: Since their conception, expressed sequence tags (ESTs) have increased in number and tissue variety, making them a valuable tool for regional gene cloning, as well as for the global effort to saturate the human transcript map. A substantial gap exists, however, between the generation of large volumes of data and their analysis, particularly with respect to gene function. The research application outlined here aims to bridge this gap and addresses the issue of tissue-specific gene expression in the human retina. To accomplish this, the information deposited in the GenBank EST database (dbEST) will be used to extract ESTs that are found only in retina libraries and no other tissue. These ESTs will be mapped in the human genome to generate a retina-specific transcript map, onto which all known mapped genetic ocular diseases will be superimposed. As a result, a substantial number of positional candidate genes for a wide range of retinopathies will be identified and offered to the scientific community through the Internet. To increase the versatility of the in silico subtraction algorithm, the investigator will subsequently construct routines that will examine large volumes of BLAST output files and determine the putative expression of any given EST through alignments with ESTs from various tissues. This will allow for a broadening of the scope of the searches to include additional tissues, such as pineal gland, searches for combinations of tissues or selecting the tissue specificity of ESTs on a ratio basis, e.g., 80 percent retina specificity. Upon completion of the computer processing of the data, all subtracted ESTs will be arrayed on a glass surface and hybridized with cDNA from retina, as well as other tissues such as brain, heart and testis. This will serve to both verify the transcription potential of the in silico subtracted EST collection and give initial biological data on the spatio-temporal message distribution for these new genes, which will be a strong base for the design of future experiments.